Self-adaptive picture cutting and fusing method and system, computer equipment and medium

Abstract

The invention provides a self-adaptive picture cutting and fusing method and system, computer equipment and a medium, and the method comprises the following steps: obtaining a video file, and extracting text content, audio content and video key frames from the video file; performing natural language analysis on the text content, and generating a semantic tag according to an analysis result; performing audio matching according to the audio content, and generating an audio tag according to a matching result in combination with an audio knowledge graph; performing content prediction according to the video key frame to generate a content label; the video file is clipped according to the semantic tag, the audio tag and the content tag, and the clipped materials are fused to obtain the target video, so that the content prediction rate is improved, the labor cost is saved, the convenience of short video application is improved, the application scene is wider, and the user experience is improved. The method is widely applied to the technical field of video processing.

Description

technical field [0001] The invention relates to the technical field of video processing, in particular to an adaptive picture cropping and fusion method, system, computer equipment and storage medium. Background technique [0002] At present, related technologies mainly target long videos (over 60 seconds) for video classification labeling. Generally, short video labeling is completed by analyzing the content of the video. It is mainly divided into the following two scenarios: First, through manual editing. , manually review the entire video content, combined with subjective judgment and understanding, and label the video with classification tags; second, through AI recognition technology, face, scene, and object recognition are performed on the frames that appear in the video content itself, and tags of corresponding categories are extracted, for example Stars, food, libraries, etc. [0003] However, due to the labor-intensive work of manually editing classification labels...

AI Technical Summary

Problems solved by technology

[0003] However, due to the labor-intensive work of manually editing classification labels, it requires editors to have high aesthetic ability and patience. Therefore, the manual labeling method often has low efficiency, slow speed, and high subjectivity of label quality , and low video frame coverage

In addition, some related technologies adopt the method of AI identification and labeling, but the method of AI identification and labeling cannot be applied to application scenarios with too many interference pictures, and the scope of application is limited

In addition, the AI ​​identification and labeling method in related technologies can only identify and label the target subject is limited, so there is still a problem of insufficient effectiveness

Images